A novel modeling approach of negative-sequence current for electrified railway traction substation. (May 2019)

• Record Type: Journal Article
• Authors: Yang, Shaobing; Li, Xue; Song, Kejian; Wu, Mingli
• Is Part Of: International journal of electrical power & energy systems. Volume 107(2019)
• Page Start: 462
• Page End: 471
• Publication Date: 2019-05
• Subjects: Ant colony optimization -- Electrified railway -- Negative-sequence current -- Probabilistic load model -- Traction load; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2018.12.004
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: Generalized and simplified probabilistic models of traction load are proposed. Number of trains, train power, and load allocation are used for NSC simulation. An improved ACO algorithm is developed for parameter identification. Abstract: To analyze the negative effects caused by unbalanced traction loads, the key is to build an efficient load model. This paper proposes a probabilistic traction load model in both generalized and simplified forms, where two decisive factors, number of trains (TN) and train power (TP), are selected as the components. Based on data analysis, Poisson and normal distributions are selected to respectively describe TN and TP. With the flexibility of stochastic approach, the model can fit various load profiles by parameter match rather than structure modification, thus it has good applicability and low complexity. Moreover, a load allocation proportion (LP) is employed to represent the unbalanced level between two traction phases, hence negative-sequence current (NSC) can be computed according to the traction transformer type, and the detailed equations and steps are given as well. Considering the correlation of TN, TP, and LP, an improved ant colony optimization is developed for parameter identification. Case studies have been implemented for model verification, parameter identification, and NSC evaluation. The simulations show that the proposed model can well describe the characteristics of both traction load and NSC.